# Drop in results upon addition of new features in random forest model

I am training a classification random forest for object detection in images. I have several features (like HoG, edge features etc) which work good enough separately. But when I train using all features together, the results don't improve. E.g. area under curve are as follows:

HoG Features: 0.90

edge Features: 0.81

Combined together: 0.86

I am using scikit-learn random forest library, # of trees = 200, information gain = 'entropy', 2 classes and I have 4000 training examples.

• Area under what curve? 200 trees seems small. How many features? – Sycorax Apr 5 '16 at 2:23
• sensitivity vs specificity. Hog features = 2500, Edge Features = 2700. Trees size kept small due to time constraints – Azhar Apr 5 '16 at 2:42
• It's plausible that your forest isn't diverse enough; try adding more trees. Moreover, by setting minimum node size to be larger, you can get calibrated probabilities while reducing training time per tree. – Sycorax Apr 5 '16 at 2:44
• Increased the tree size to 100, no improvement – Azhar Apr 5 '16 at 20:18
• If you had 200 trees before, 100 trees is a decrease. – Sycorax Apr 5 '16 at 20:19

You could try increasing max_features to combat this.